On Wed, 2005-02-23 at 13:19 +0300, Riad Souissi wrote:
> Hello,
> 
> In case the HTTPClient is set up to preserve Cookies, I have the
> following problem:
> 
> - Sometimes, HTTPClient presents a cookie which is not valid anymore in
> the target HTTP server (because it timed out, or the server cleaned up
> some sessions or the server restarted, etc).
> - The HTTP server returns HTTP 401 error code
> - The HTTPClient re-attemps to authenticate, but still re-use the same
> cookie id.
> - All subsequent executes that involve the HTTPConnection which has this
> cookie will fail, until the Cookies are expired in the HTTPClient.
> - I thought about catching the 401 code on top of the HTTPClient and
> cleanup the Cookie my self. The problem is that I do not know which
> Cookie ID has a problem and from which HTTPConnection object.
> 
> My guess/suggestions is that the HTTPClient could be made to re-set the
> Cookie from the HTTPConnection object it is currently using when it
> receives HTTP 401 code and it had already provided the correct
> Authentication credentials before (which is my case since I use BASIC
> and setAuthenticationPreemptive=true).
> 

Riad,

For what it's worth, in my opinion this assumption is application
specific. The stock version of HttpClient can't (and should not) know
which cookie represents a session token, and should make no attempts to
invalidate it. It should be trivial to implement a custom handler on top
of HttpClient that clears HttpState upon receipt of 401 status code

Oleg

> Any idea/suggestions/comments ?
> 
> Thanks
> riad
> 
> 
> DiscalimerThe information in this email and in any files transmitted with it,
> is intended only for the addressee and may contain confidential and/or 
> privileged material.
> Access to this email by anyone else is unauthorized. If you receive this in 
> error,
> please contact the sender immediately and delete the material from any 
> computer.
> If you are not the intended recipient, any disclosure, copying, distribution 
> or
> any action taken or omitted to be taken in reliance on it, is strictly 
> prohibited.
> Statement and opinions expressed in this e-mail are those of the sender, and 
> do not
> necessarily reflect those of STC.


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to